AIO Papers 7 the Last Erechtheion Building Accounts

______

The Last Erechtheion Building Accounts

S. D. Lambert

with a contribution by John D. Morgan

AIO Papers no. 7 January 2016 Links added: May 2016 AIO Papers

Published by Attic Inscriptions Online, 97 Elm Road, Evesham, Worcestershire, WR11 3DR, United Kingdom.

Editor: Dr. S. D. Lambert

AIO Advisory Board: Professor P. J. Rhodes (Durham) Professor J. Blok (Utrecht) Dr. A. P. Matthaiou (Athens) Mr. S. G. Byrne (Melbourne) Dr. P. Liddel (Manchester)

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Attic Inscriptions Online, or as expressly permitted by law, or under terms agreed with the appropriate reprographic rights organisation. Enquiries concerning reproduction should be sent to Dr. S. D. Lambert at the above address or via the contact given at www.atticinscriptions.com.

ISSN 2054-6769 (Print) ISSN 2054-6777 (Online)

THE LAST ERECHTHEION BUILDING ACCOUNTS1

S. D. Lambert with a contribution by John D. Morgan

The main body of the building accounts of the Erechtheion are in Attic script: IG I3 474-475, of 409/8 BC, and IG I3 476, of 408/7 (and possibly 407/62). There are also two sets of accounts in Ionic script:3

1. The most substantial comprises the two joining fragments, Caskey XXVII (= fr. a below) – XXVIII (= fr. b below) (pp. 418-420, a better text than IG II2 1654). Lambert 2000 (SEG 50.69) showed that these fragments belong with IG I3 478 (= fr. c below), as E. Schweigert had argued (Hesp. 7, 1938, 268-269), and that, correctly read, several of the workers named on the fragments can be identified with workers named on the accounts in Attic script. This undermines the idea, espoused in IG I3, that these accounts date to some time in the early fourth century, and suggests rather that they belong in a group with the other surviving accounts. In XXVIII (fr. b), l. 28, there is reference to work on parts of the temple damaged by fire. This is most naturally taken as repair work following the acropolis fire of 406 BC (Xen. Hell. 1.6.1). Since, as we shall see, no. 2 below (IG I3 477) dates most comfortably to the third and fourth prytanies of 405/4, 406/5 is a good candidate for the year of fr. b, with fr. a belonging either to the same year or to 407/6. In 2000 I did not supply a continuous text and my revisions, and

1 I am very grateful to John Morgan for his advice on the date of IG I3 477 and for his permission to include in this paper his discussion of its broader calendrical implications. Caskey: L. D. Caskey, “The Inscriptions”, ch. 4 of J. M. Paton ed., The Erechtheum (Cambridge, Mass., 1927); Dinsmoor 1932: W. B. Dinsmoor, “The Burning of the Opisthodomos at Athens”, AJA 36, 143-172 (cf. SEG 10.299); Lambert 2000: S. D. Lambert, “The Erechtheum Workers of IG II2 1654”, ZPE 132, 157-160 (SEG 50.69). 2 Cf. Lambert 2000, 160 n. 16. 3 I exclude IG I3 479 (= SEG 10.301), which is too fragmentary for confident identification as an Erechtheion account.

the subsequent discussion of the name in l. 2, have yet to be incorporated in the PHI database. For the ease of reference of users of AIO I supply a fresh text below. It is not based on any new primary work, but merely represents the textual status quo following my 2000 article and reactions to it.

col. 1 col. 2 ․ ․ . ․8․ ․ ․ ․ Ε̣ – – – – – – – – – – – – – – – – stoich. fr. a [.]Α̣ ΔΜΟΝΑΝΑ[– – – – – – – – – ἐμ Με]– [λ]ίτ(ηι) οἰκο͂ν(τα) Δ– – – – – – – – – – – – [.]σι οἰκο͂ν(τα) Δ– – – – – – – – – – – – – 5 [.]Δ· Ἀρίστωνα Ἀλω[πε(κῆσι) οἰκ( ντα) –] [.] ἐμ Μελί(τηι) οἰκ( ντα) Ńł – – – – – – – – [Μ]υννίωνα ἐμ Μελ̣[ί(τηι) οἰκ( ντα) – – – –] [Σ]ῶ̣κλον Ἀλωπε(κῆσι) ο[ἰκ( ντα) – – –] [.] ἐτιμήθη ΔΔŃ[– – – – – – – – – – – τ]– 10 [ῆ]ι βολῆι ἐς [– – – – – – – – – – τῶι δή]– [μ]ωι δoξαν – – – – – – – – – – – – – – [Ἀ]γγελῆθ[εν – – – – – – – – – – – – – [.]ΔŃł· Φ⟨υ⟩[ρόμαχ- Kηφισι- – – – – –] [.]ιτον. [– – – – – – – – – – – – – – – –] 15 [.]γραφο[– – – – – – – – – – – ἐν Κολ?]– λυ(τῶι?) οἰκῶ[ν – – – – – – – – – – – –] ἐς τὸ β[– – – – – – – – – – – – – – – Β]– ατῆθ̣[εν – – – – – – – – – – – – – – – [.]της – – – – – – – – – – – – – – – – 20 [.]ον – – – – – – – – – – – – – – – –

[.]πρ – – – – – – – – – – – – – – – – [․]Ι – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – 25 . . . . . ․11․ ․ ․ ․ ․ Ι – – – – – – – – – – fr. b [. .]I[. . .6. . .]ΑΡΧΟΝ[– – – – – – – –] [τρ]ί[τηι τ]ῆς πρυτα[νέας – – – – – –] [․ τ]ο͂ τ̣ε ν̣εὼ τὰ κεκα[υμένα – – – –] [ψη]φισαμένης τῆς β[ολῆς – – – – – –] 30 [․ ․ ․] ἐγ Κολλ(υτῶι) οἰκῶ(ν) ΔΔ[– – – τῆς πρυ]– [τα]νέας τo ἐργαστη[ριο– – – – – – – – –] [Δ]ρωπίδης ἐν Μελί(τηι) οἰκ̣[ῶν – – – – – τ]– ῆς πρυτανέας μισθώμ[ατα – – – – – Θ]– ρασωνίδηι Κικυννεῖ – – – – – – – – – 35 [.]να̣ίος τὸς ἐπὶ τῶν τοί[χων – – – – –] [.]ς κατὰ τὸ Πανδρόσειο[ν – – – – δρ]– [α]χμῶν ἕκαστον μισθω[.]Ο/[ – – – – –] [․]ńΔłłłΙΙΙΙΙ· ἑβδόμηι τῆ[ς πρυτανέας – – τὸς ὀ] [χετ?]ὸς τὸς καινὸς ἐξεργ[ά]ζ[ομένωι – –] 40 [δραχμ]ῶν τὴν εἰκοσίπο[δ]ί[αν – – – – – –] ․ ․ . 6․ ․ ․ Δ̣ŃΙΙΙ ἐς τὴν ἄλει[ψιν – – – – –] [․ ․ .5․ ․ π]α̣ρὰ Σατύρας ἐ Σκ̣[αμβω(νίδαις) οἰκ(όσης) – –] [․ ․ ․ ․7․ ․ . χά?]λ̣ιξ ̣ ἐς τὸς ὀχε[τός – – – –] ․ ․ ․ ․ .10․ ․ ․ ․ . ł παλιμ– – – – – – – – – 45 ․ ․ ․ ․ . ․12․ ․ ․ ․ . ․ ΣΤΩ̣ – – – – – – – – –

lacuna – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – ΙΙΙ, Πρέ̣ποντ– fr. c 50 οἰ[κ ντι – – – – – – – – – – – – – – ο]- [ι – – – – – – – – – – – – – – –]ΙΙΙ, Σισύφωι ἰκ [ντι – – – – – – – – – – – – – – – –] [– – – – – – – – – – – – – – –] τῆς πρυτανε– Παρμ[ένοντι Λαόσσο – – – – – – – –] [ίας – – – – – – – – – – – – – ]στωι ΔΔ Ἱ̣ερ– κι– – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – –

The position of fr. c in relation to fragments a and b is uncertain. The readings of a + b are those recorded by Caskey, revised by Lambert 2000, the readings of fr. c are those of Schweigert, confirmed, or in the case of l. 49, revised by Lambert 2000. Supplements are by Koehler, 5, 38-39, 39 fin., 40, 42 Caskey, 8, 13, 17-18, 43 Lambert, 15-16, 32 in. Wilamowitz, 41 Preuner, fr. c Schweigert. 1-2 [Εὔ|κ]α̣δμον Ἀνα[καιέα], paternal grandfather of the Eukadmos of Anakaia who was anagrapheus under the oligarchy in 319/8 BC (Agora XVI 101) and the sculptor whose student Androsthenes carved the figures in the temple of Apollo at Delphi (Paus. 10.19.4), probably in 327 BC (J. Bousquet, BCH 108 [1984], 695-698), J. D. Morgan, AJA 106 (2002), 296 (SEG 50.69). [Φ|ρά]δμονα previous eds. A. Corso, NAC 30 (2001), 53-71 (SEG 51.2355) suggests that he is the Phradmon of Argos, sculptor in bronze ca. 420 BC (Pliny, NH 34.49 and 53; Paus. 6.8.1), but one would expect an Argive to be identified in this inscription as a metic, by Attic deme of residence, and Να- are not the initial letters of any Attic deme. Also possible: Phradmon (slave) of Na-. 5 Perhaps identifiable with Ἀριστ-, named in the company of sculptors/stone-workers at IG I3 476, 402. 7 Identifiable with the sculptor, Mynnion of Agryle, IG I3 476, 169-70, listed there, as here, immediately before Soklos, supposing that, in the meantime, he had changed deme of residence. 8 Identifiable with the sculptor Soklos resident in Alopeke, IG I3 476, 173. 11-12 Lambert 2000, 158 tentatively suggested that this may have been a reference to Καλλίας Ἀγγελῆθεν, eponymous archon in 406/5. 13 ΦΨ- stone. Identifiable as Phyromachos of Kephisia, sculptor at IG I3 476, 144, 159, 175, 419. 14 Lambert 2000 suggested [Θευγ|ε]ίτονο̣[ς or [Θευγ|ε]ίτονι, ̣ ? = the stone worker Theugeiton of Piraeus at IG I3 476, 99, 217, 325. 15-16 Φ]|λυ οἰ

17 β[ολευτήριον dub. Koe. 26 [ἐπ|ὶ Δ]ι[οφάντο] ἄρχοντος (395/4) Koe., [ἐπ]ὶ [Καλλίο] ἄρχον[τος (406/5) or [ἐπ]ὶ [Ἀλεξίο] ἄρχον[τος (405/4) or -θε]ν̣ [καὶ συν]αρχόν[των] Caskey and Dinsmoor pp. 152-60. 30-33 Structural work on a workshop is recorded at IG I3 475, 263, but Caskey/Dinsmoor’s τὸ ἐργαστή[ριον οἰκοδομήσαντι] is uncertain, as are their restorations of the days in prytany as τετάρτηι (30), πέμπτηι (32). 39 fin. τριῶν Dinsmoor. 43 For χάλιξ cf. Thuc. 1.93.5, IG I3 387, 44. On the ὀχετοί in the Erechtheum, Caskey p. 356 (inconclusive). 46-47 Identifiable as the metic Prepon resident in Agryle, listed as a stone-cutter at IG I3 476, 14, 24 etc. 47 Identifiable as the metic goldsmith Sisyphos resident in Melite, IG I3 476, 57-58. 52 Identifiable as the Parmenon slave of Laossos, IG I3 476, 77, 225, 308.

2. IG I3 477 (= IG II2 1655 = Caskey XXIX).

– – – – – – – – – – – – – – – – – – – – – – – – εσ. . εσ łΙΙΙ v χάρτης [εἰς ἐγγραφὴν ἀντιγράφ]- stoich. 35 ων łłłΙΙΙ vvv λήμματο[ς καὶ σύμπαντος ἀναλώ]- ματος κεφάλαιον ŅΗΗΗ – – – – – – – – – – ἐπὶ τῆς Πανδιονίδος τετ[άρτης πρυτανευόσ]- 5 ης λῆμμα παρὰ ταμιῶν τῆς [θεο͂ ․ . ․ ․ . ․11․ ․ ․ ․ ․ Π]- ροβαλισίο καὶ συναρχόν[των ἕκτηι(?) καὶ δεκά]- τηι τῆς πρυτανείας∶ Χ v [v εἰς θυσίαν τοῖς Χα]- λ̣κείοις τῆι Ἀθηναίαι [․ ․ ․ ․ ․ ․ . . ․17․ ․ ․ ․ ․ ․ ․ ․] [. ․ ․ ․8․ ․ ․ ․]ΛΤΛΛ․ ․Σ – – – – – – – – – – – –

Readings and restorations, which mostly follow the first edition of Lolling, Δελτ. Ἀρχ. 1888, 118 no. 2, are as recorded in IG I3, except ll. 7- 8. 1 ς Bannier ap. IG II2 1655, cf. IG I3 476, 188-190. 6 Dinsmoor 1932. 7-8 R. Parker, Polytheism and Society at Athens (2005), 464; Ἀν]|ακείοις eds.

If Parker’s attractive suggestion is correct, it enables some progress to be made on the determination of the year (407-405 in IG I3), as the date of the Chalkeia was the last day of Pyanopsion, Πυανοψιῶνος ἕνηι καὶ νέαι (Harp. s.v. Χαλκεῖα etc., see Parker, 464), and the date of the sacrifice in this inscription should therefore have been Pryt. IV 16 (?) = Pyan. ult.. John Morgan kindly advises as follows:

“The following analysis is based on the four assumptions that (i) the festival to be restored in lines 7-8 of IG I3 477 is not the Anakeia but the Chalkeia, (ii) that the payment was made on, or within a few days of, the celebration of the Chalkeia on the last day of Pyanopsion, (iii) that the archon’s calendar was not at that time retarded by the insertion of several embolimoi days, and (iv) that the stoichedon order, with 35 letter- spaces per line, was maintained in the missing right part of this inscription. If any of these four assumptions is incorrect, the analysis and conclusions below may turn out to be invalid.”

“With 35 letters per line, lines 4-8 can be restored as either

ἐπὶ τῆς Πανδιονίδος τετ[άρτης πρυτανευόσ]- 5 ης λῆμμα παρὰ ταμιῶν τῆς [θεο͂ ․ . ․ ․ . ․11․ ․ ․ ․ ․ Π]- ροβαλισίο καὶ συναρχόν[των ἕκτηι καὶ δεκά]- τηι τῆς πρυτανείας∶ Χ v [v εἰς θυσίαν τοῖς Χα]- λ̣κείοις τῆι Ἀθηναίαι [․ ․ ․ ․ ․ ․ . . ․17․ ․ ․ ․ ․ ․ ․ ․] or

ἐπὶ τῆς Πανδιονίδος τετ[άρτης πρυτανευόσ]- 5 ης λῆμμα παρὰ ταμιῶν τῆς [θεο͂ ․ . ․ ․ . ․11․ ․ ․ ․ ․ Π]- ροβαλισίο καὶ συναρχόν[των μιᾶι καὶ εἰκοσ]- τηι τῆς πρυτανείας∶ Χ v [v εἰς θυσίαν τοῖς Χα]- λ̣κείοις τῆι Ἀθηναίαι [․ ․ ․ ․ ․ ․ . . ․17․ ․ ․ ․ ․ ․ ․ ․]”

“If this document were to belong to a time within the period of 10 tribes when the archon’s year and the prytany year were coterminous, with 36 days in each of the first 4 prytanies and 35 days in the remaining 6 prytanies in an ordinary year (4 x 36 + 6 x 35 = 354),4 and 39 days in each of the first 4 prytanies and 38 days in the remaining 6 prytanies in an intercalary year (4 x 39 + 6 x 38 = 384),5 the 16th day of the 4th prytany would be the (3 x 36) + 16 = 124th day in an ordinary year or the (3 x 39) + 16 = 133rd day in an intercalary year, and the 21st day of the 4th prytany would fall 5 days later on the 129th day in an ordinary year or the 138th day in an intercalary year. With lunar months averaging about 29½ days, the last day of Pyanopsion, the 4th month of the year, would normally fall on or about the 118th day of the year in an ordinary year or an intercalary year, unless in an intercalary year an intercalary month was inserted some time before Pyanopsion,6 in which case the last day of Pyanopsion would fall on or about the 147th or 148th day of the year. In none of these cases would there be a close coincidence of the 16th or the 21st day of the 4th prytany with the last day of Pyanopsion. Hence, given the above assumptions, it is reasonable to infer that this document does not belong to the time when the archon’s year and the prytany year were coterminous, but rather to the preceding time in the last decades of the 5th century BC when the prytany year consisted of 365 or 366 days, with 37 days in the first 5 or 6 prytanies and 36 days in the remaining 5 or 4 prytanies, respectively. The logistai inscription (IG I2 324 = IG I3 369) provides good contemporary epigraphical

4 [Aristotle], Athenaion Politeia 43.2. 5 As was reasonably hypothesized by W.K. Pritchett and O. Neugebauer, The Calendars of Athens (Cambridge, Mass., 1947), 36-37. 6 E.g. the intercalary month was sometimes inserted after Hekatombaion, as is attested in the 420s BC by IG I3 78 and in 227/6 by Agora XV 120 = IG II3 1, 1139 and in 225/4 by IG II2 838 = IG II3 1, 1146 and Agora XVI 224 = IG II3 1, 1147.

evidence that the 4 prytany years from the summer of 426 to the summer of 422 BC consisted of 366 days each,7 but the relatively scrappy literary evidence thereafter seems to indicate that between 422 and 411 BC five prytany years were 366 days long and 6 others only 365 days long,8 so both possibilities should be taken into account when considering the year in which IG I3 477 belongs.”

“If for no other reason, the fact that in Euktemon’s archonship (408/7) Pandionis held the 8th prytany (as is shown by lines 183-184 of the Erectheion account IG I3 476) shows that 408/7 cannot be the year of IG I3 477, when Pandionis held the 4th prytany.”

“The calendar equations on the lower portion of the reverse face of the Choiseul Marble (IG I3 377), e.g. Prytany I.20 = Hekatombaion 20 and Prytany II.1 = Metageitnion 8, indicate that the archon’s year and the prytany year 407/6 began on the same day, with 37 days in the first prytany and 30 days in Hekatombaion. Hence in this year the 16th day of the 4th prytany was the (3 x 37) + 16 = 127th day of the prytany year, and the 21st day of the 4th prytany was the 132nd day of the prytany year. Neither of these would be close to the celebration of the Chalkeia on or about the 118th day of the archon’s year, which in this year began on the same day as the prytany year. Hence it seems that IG I3 477 should not be attributed to the Athenian year 407/6.”

“With 354 or 355 days in the archon’s year 407/6 and 365 or 366 days in the prytany year 407/6, these correspondences would be shifted by 10 or 11 or 12 days so that the 127th and 132nd days of the prytany year 406/5 would fall around the 137th to 139th days and 142nd to 144th days of the archon’s year 406/5. Again, there would not be a close correspondence with the celebration of the Chalkeia on or about the 118th day of the archon’s year. Nor would there be a close correspondence if an intercalary month of 29 or 30 days had been inserted in the archon’s calendar in 407/6 or some time before Pyanopsion in 406/5, in which case the Chalkeia would have been shifted 29 or 30 days later. Hence it seems that IG I3 477 should not be attributed to the Athenian year 406/5.”

7As Pritchett consistently contended in The Calendars of Athens (1947), 94-105, and in many works thereafter, with the concurrence of Mabel Lang (“The Abacus and the Calendar”, Hesp. 33 (1964), 146-167, and 34 (1965), 224-247) and eventually Meritt himself (M. Lang and B.D. Meritt, “A New Text of the Logistai Inscription”, CQ 18 (1968), 84-94). 8 See B.D. Meritt, The Athenian Year (Berkeley, 1961), 202-212, for a discussion of this evidence.

“Assuming that one of the two archon years 407/6 and 406/5 was intercalary, so that these two years contained about 354 + 384 = 738 days, and that the prytany years 407/6 and 406/5 contained 365 + 365 = 730 days or 365 + 366 = 731 days or 366 + 366 = 732 days, we can obtain a close correspondence of the 16th day of the 4th prytany in 405/4 with the celebration of the Chalkeia on the last day of Pyanopsion, the 118th day of the archon’s year. Counting forward from the calendar equation

Pr. I.1 = Hekatombaion 1 in the summer of 407 BC by 730 days yields the calendar equation

Pr. I.1 ≈ Skirophorion 22 in the summer of 405 BC.

(Note that 2 x 365 = 730 = 355 + 354 + 21.) With 8 or 7 days remaining in Skirophorion, the 16th day of the 4th prytany would fall (3 x 37) + 15 = 126 days later on or about the 118th or 119th day of the archon’s year, within a day or so of the celebration of the Chalkeia on the last day of Pyanopsion.”

“Hence this analysis, which is based on the 4 assumptions mentioned above, suggests that IG I3 477 should probably be attributed to the Athenian year 405/4.”

“If this analysis is correct, it indicates that the independent prytany year was abolished not in the summer of 407 BC, as B.D. Meritt9 and W.K. Pritchett10 contended, but some years later, perhaps in or around 404/3 BC, the year of anarchia.11 In any case, it should be observed that the “extreme coincidence” in equating the first day of the first prytany with the first day of Hekatombaion in the summer of 407 BC is a natural consequence of the equation of the first day of what would have been the first prytany with the 14th day of Skirophorion in the

9 B.D. Meritt, “Senatorial and Civil Years in Athens”, CP 25 (1930), 236-243; The Athenian Year (Berkeley, 1961), 212-215; “Athenian Calendar Problems”, TAPhA 95 (1964), 204-212. 10 W.K. Pritchett, The Choiseul Marble (Berkeley, 1970), 33-34; “The Choiseul Marble: A Palimpsest with Graffiti”, BCH 101 (1977), 28-29. 11 As Meritt had supposed in his first book, The Athenian Calendar in the Fifth Century (Cambridge, Mass., 1928), 124-126.

summer of 411 BC.12 With 366 days in each of the 4 prytany years 411/0, 410/09, 409/8, and 408/7, the first day of the first prytany in the summer of 407 BC would fall 4 x 366 = 1464 days after the 14th of Skirophorion in the summer of 411 BC. 3 ordinary years and 1 intercalary year, with a total of 49 lunar months averaging 29.53059 days, would comprise 1447 days, but in fact from the astronomical new moon on 23 June 411 BC (which presumably was the ἕνη καὶ νέα of Thargelion in 412/1 BC) to the astronomical new moon on 10 June 407 BC (which presumably was the ἕνη καὶ νέα of Thargelion in 408/7 BC) is 1461 – 13 = 1448 days, which is less than 1464 days by 16 days. The 16th day after the 14th of Skirophorion (if it was a hollow month of 29 days, as is suggested by the well-preserved calendar equations in Skirophorion on the upper portion of the reverse face of the Choiseul Marble13 and by the dates and times of astronomical new moons in the summer of 407 BC14) would be the first day of Hekatombaion. Hence this “extreme coincidence” is not, as Pritchett supposed, evidence for tampering with either the archon’s calendar or the prytany calendar near the end of 408/7 to make these calendars coterminous starting in 407/6 and thereafter.” For Morgan’s argument on this point see also AIO Papers 5 [2014], p. 3.

On this basis IG I3 477 would be the latest extant Erechtheion account, belonging probably to the year after no. 1 above. It would seem quite appropriate that provision should be made for a celebratory sacrifice to Athena at the Chalkeia by the workers on the Erechtheion at or near the completion of their work.

12 [Aristotle], Athenaion Politeia 32.1. 13 The well-preserved calendar equations Pr. X.12 = Skirophorion 5 and Pr. X.23 = Skirophorion 16 on the upper portion of the reverse face of the Choiseul Marble (IG I3 377) are consistent with the 10th prytany having 36 days (as was normal in the independent prytany calendar in the 5th century BC) and Skirophorion being a hollow month of 29 days, with no irregular adjustments made in either the prytany calendar or the archon’s calendar. 14 Fred Espenak’s Astropixels website shows astronomical new moons in 407 BC (-406 in astronomical notation, because there is no year 0 B.C. or A.D.) on June 10 at 03:24 Universal Time and on July 9 at 10:36 Universal Time; local time at Athens would be about 90 minutes later. The elapsed time between these events is 29 days and about 7 hours, which is closer to 29 days than to 30 days. New crescent moons were probably visible on the evenings of June 11 and July 10, and the elapsed time is 29 days.